Non-invasive, regional assessment of lung function has the potential to markedly enhance early diagnosis and management of disease, to accelerate drug discovery and trials, and to provide fundamental insights into the root causes of disease. In this proposal, we aim to address a fundamental question of emphysema pathogenesis thorough a systematic study of lung function and structure using techniques developed over the first funding cycle. These techniques include measurements of pulmonary ventilation and perfusion, alveolar oxygen tension, and a sensitive measure of lung microstructure using hyperpolarized magnetic tracers, as well as several well-established methods in pulmonary medicine. The proposed research is designed specifically to corroborate or refute a molecular model of emphysema progression that is currently under widespread debate. Despite extensive investigation, it is still not clear which of the numerous and varied presentations of emphysema corresponds to the fundamental cause of the disease, or even if one principal cause can be identified. Nonetheless, a large body of current research suggests that altered expression of one or more mediators of angiogenesis may be the first step in the disease. This alteration degrades the capillary endothelium's ability to repair itself, resulting in cell death, reduced perfusion, and ultimately tissue destruction. Although in vivo imaging methods cannot replace the powerful molecular techniques in use today, they can provide a more complete picture of the action of disease or therapy on the body. In this study, we first demonstrate a comprehensive, self-consistent, and validated set of pulmonary measurements and evaluate sensitivity to disease. We then apply these measurements to three animal models of emphysema in order to evaluate time ordering and spatial correlation of the various aspects of pulmonary dysfunction. By comparing these measurements to testable predictions of the disease model, we hope to elucidate the first stages of this devastating disease. [unreadable] [unreadable] [unreadable]